Does martian soil release reactive halogens to the atmosphere?

Detailed statistical examination of Cl, Br, and S distributions, in martian soil profiles at Gusev Crater and Meridiani Planum, indicates decreasing Br abundance and weakening Br–S association towards the surface. All three elements decrease towards the surface in the order Cl < S < Br. Furthermore, Br variability decouples from potential cations such as Mg at the surface relative to the subsurface. These observations support a relative loss of surficial Br compared to S and Cl, all highly mobile elements in aqueous environments. We propose that Br may have converted preferentially to gas phases (e.g., BrO), driven either by UV photolysis or by chemical oxidants. Such volatilization pathways may in turn impart a global signature on Mars by acting as controls on oxidants such as ozone and perchlorates. S/Cl mass ratios vary with depth (∼4–5 in the subsurface; 1.8–3.6 on the surface) as well, with a strong correlation of S and Cl near the surface but more variable at depth, consistent with differential vertical mobility, but not volatilization of Cl. Elevated S/Cl in subsurface soil also suggests that the ratio may be higher in bulk soil – a key repository of martian geologic and climatic records – than previously thought.     

Experimental constraints on the evaporation of partially oxidized acid-sulfate waters at the martian surface

Isothermal evaporation experiments were carried out on an acidic (pH 2), partially oxidized (Fe²⁺/Fe_T ∼0.5) brine with a cation composition consistent with derivation from the chemical weathering of martian basalt. During evaporation, the brine composition evolved to a highly acidic (un-scaled pH −1.3) Mg-Fe-SO₄-Cl brine depleted in Ca, Al and K. Evaporite minerals identified throughout the course of the experiment include (in order of crystallization): gypsum, Mg-rich voltaite, (Mg_0.7, )SO₄·7H₂O and rhomboclase. The solid solution compositions of voltaite and (Mg_0.7, )SO₄·7H₂O, although uncommon in analogous environments on Earth, result from the distinct chemistry of evaporating martian surface fluids. Analysis of brine compositions with available thermodynamic models indicates that, although gypsum and rhomboclase precipitate at equilibrium saturation, kinetic controls on the precipitation of copiapite-group minerals affect the subsequent sulfate mineralogy and evolving chemistry of the entire system. In addition, geochemical simulations of the experimental evaporation process suggest that the appearance of voltaite and rhomboclase indicate a “metastable” evaporation pathway for martian fluids whereby bilinite and copiapite-group minerals did not form despite thermodynamic saturation. Comparison of the experimentally-produced assemblage to available observations of saline minerals at the martian surface represents a step toward systematically characterizing evaporite mineralogy as a function of Fe-oxidation in the initially dilute fluid. Deconvolving the complexity of Fe-sulfate formation in martian environments ultimately will help to exploit the sensitivity of these minerals to pH and redox conditions present at the ancient martian surface.     

Role of subducted sediments in island-arc magmatism: constraints from REE patterns

Continental sedimentary rocks of post-Archean age are characterized by europium depletion relative to the other REE. Typical values for Eu/Eu* are 0.65±0.05 (where Eu* is the theoretical value for no chondrite-normalized Eu anomaly).Basalts and andesites from island-arc suites rarely display significant europium anomalies. Calculations indicate that the maximum amount of sediment that can be admixed without producing a signature of Eu depletion is: (a) 10% for a MORB source; (b) 1% for primitive mantle or a single-stage depleted mantle, and (c) 0.1–0.3% for a two-stage depleted mantle.     

Stratigraphy and sedimentology of a dry to wet eolian depositional system, Burns formation, Meridiani Planum, Mars

Outcrop exposures of sedimentary rocks at the Opportunity landing site (Meridiani Planum) form a set of genetically related strata defined here informally as the Burns formation. This formation can be subdivided into lower, middle, and upper units which, respectively, represent eolian dune, eolian sand sheet, and mixed eolian sand sheet and interdune facies associations. Collectively, these three units are at least 7 m thick and define a “wetting-upward” succession which records a progressive increase in the influence of groundwater and, ultimately, surface water in controlling primary depositional processes.The Burns lower unit is interpreted as a dry dune field (though grain composition indicates an evaporitic source), whose preserved record of large-scale cross-bedded sandstones indicates either superimposed bedforms of variable size or reactivation of lee-side slip faces by episodic (possibly seasonal) changes in wind direction. The boundary between the lower and middle units is a significant eolian deflation surface. This surface is interpreted to record eolian erosion down to the capillary fringe of the water table, where increased resistance to wind-induced erosion was promoted by increased sediment cohesiveness in the capillary fringe. The overlying Burns middle unit is characterized by fine-scale planar-laminated to low-angle-stratified sandstones. These sandstones accumulated during lateral migration of eolian impact ripples over the flat to gently undulating sand sheet surface. In terrestrial settings, sand sheets may form an intermediate environment between dune fields and interdune or playa surfaces. The contact between the middle and upper units of the Burns formation is interpreted as a diagenetic front, where recrystallization in the phreatic or capillary zones may have occurred. The upper unit of the Burns formation contains a mixture of sand sheet facies and interdune facies. Interdune facies include wavy bedding, irregular lamination with convolute bedding and possible small tepee or salt-ridge structures, and cm-scale festoon cross-lamination indicative of shallow subaqueous flows marked by current velocities of a few tens of cm/s. Most likely, these currents were gravity-driven, possibly unchannelized flows resulting from the flooding of interdune/playa surfaces. However, evidence for lacustrine sedimentation, including mudstones or in situ bottom-growth evaporites, has not been observed so far at Eagle and Endurance craters.Mineralogical and elemental data indicate that the eolian sandstones of the lower and middle units, as well as the subaqueous and eolian deposits of the Burns upper unit, were derived from an evaporitic source. This indirectly points to a temporally equivalent playa where lacustrine evaporites or ground-water-generated efflorescent crusts were deflated to provide a source of sand-sized particles that were entrained to form eolian dunes and sand sheets. This process is responsible for the development of sulfate eolianites at White Sands, New Mexico, and could have provided a prolific flux of sulfate sediment at Meridiani. Though evidence for surface water in the Burns formation is mostly limited to the upper unit, the associated sulfate eolianites provide strong evidence for the critical role of groundwater in controlling sediment production and stratigraphic architecture throughout the formation.     

Provenance and diagenesis of the evaporite-bearing Burns formation, Meridiani Planum, Mars

Impure reworked evaporitic sandstones, preserved on Meridiani Planum, Mars, are mixtures of roughly equal amounts of altered siliciclastic debris, of basaltic provenance (40 ± 10% by mass), and chemical constituents, dominated by evaporitic minerals (jarosite, Mg-, Ca-sulfates ± chlorides ± Fe-, Na-sulfates), hematite and possibly secondary silica (60 ± 10%). These chemical constituents and their relative abundances are not an equilibrium evaporite assemblage and to a substantial degree have been reworked by aeolian and subaqueous transport. Ultimately they formed by evaporation of acidic waters derived from interaction with olivine-bearing basalts and subsequent diagenetic alteration. The rocks experienced an extended diagenetic history, with at least two and up to four distinct episodes of cementation, including stratigraphically restricted zones of recrystallization and secondary porosity, non-randomly distributed, highly spherical millimeter-scale hematitic concretions, millimeter-scale crystal molds, interpreted to have resulted from dissolution of a highly soluble evaporite mineral, elongate to sheet-like vugs and evidence for minor synsedimentary deformation (convolute and contorted bedding, possible teepee structures or salt ridge features). Other features that may be diagenetic, but more likely are associated with relatively recent meteorite impact, are meter-scale fracture patterns, veins and polygonal fractures on rock surfaces that cut across bedding. Crystallization of minerals that originally filled the molds, early cement and sediment deformation occurred syndepositionally or during early diagenesis. All other diagenetic features are consistent with formation during later diagenesis in the phreatic (fluid saturated) zone or capillary fringe of a groundwater table under near isotropic hydrological conditions such as those expected during periodic groundwater recharge. Textural evidence suggests that rapidly formed hematitic concretions post-date the primary mineral now represented by crystal molds and early pore-filling cements but pre-date secondary moldic and vug porosity. The second generation of cements followed formation of secondary porosity. This paragenetic sequence is consistent with an extended history of syndepositional through post-depositional diagenesis in the presence of a slowly fluctuating, chemically evolving, but persistently high ionic strength groundwater system.     

Minimum Acceptance Criteria for Geostatistical Realizations

Geostatistical simulation is being used increasingly for numerical modeling of natural phenomena. The development of simulation as an alternative to kriging is the result of improved characterization of heterogeneity and a model of joint uncertainty. The popularity of simulation has increased in both mining and petroleum industries. Simulation is widely available in commercial software. Many of these software packages, however, do not necessarily provide the tools for careful checking of the geostatistical realizations prior to their use in decision-making. Moreover, practitioners may not understand all that should be checked. There are some basic checks that should be performed on all geostatistical models. This paper identifies (1) the minimum criteria that should be met by all geostatistical simulation models, and (2) the checks required to verify that these minimum criteria are satisfied. All realizations should honor the input information including the geological interpretation, the data values at their locations, the data distribution, and the correlation structure, within acceptable statistical fluctuations. Moreover, the uncertainty measured by the differences between simulated realizations should be a reasonable measure of uncertainty. A number of different applications are shown to illustrate the various checks. These checks should be an integral part of any simulation modeling work flow.     

Application of the Pitzer ion interaction model to isopiestic data for the Fe2(SO4)3-H2SO4-H2O system at 298.15 and 323.15 K

Recent isopiestic studies of the Fe₂(SO₄)₃-H₂SO₄-H₂O system at 298.15 K are represented with an extended version of Pitzer’s ion interaction model. The model represents osmotic coefficients for aqueous {(1 − y)Fe₂(SO₄)₃ + yH₂SO₄} mixtures from 0.45 to 3.0 m at 298.15 K and 0.0435 ? y ? 0.9370. In addition, a slightly less accurate representation of a more extended molality range to 5.47 m extends over the same y values, translating to a maximum ionic strength of 45 m. Recent isopiestic data for the system at 323.15 K are represented with the extended Pitzer model over a limited range in molality and solute fraction. These datasets are also represented with the usual “3-parameter” version of Pitzer’s model so that it may be incorporated in geochemical modeling software, but is a slightly less accurate representation of thermodynamic properties for this system. Comparisons made between our ion interaction model and available solubility data display partial agreement for rhomboclase and significant discrepancy for ferricopiapite. The comparisons highlight uncertainty remaining for solubility predictions in this system as well as the need for additional solubility measurements for Fe³⁺-bearing sulfate minerals. The resulting Pitzer ion interaction models provide an important step toward an accurate and comprehensive representation of thermodynamic properties in this geochemically important system.     

Geochemistry of loess,continental crustal composition and crustal model ages

Isotopic, major and trace element studies of loess deposits from America, China, Europe and New Zealand show general uniformity of composition. Silica, Zr and Hf are enriched relative to estimates of bulk composition of the upper continental crust. The REE data are indistinguishable from those of average shales, confirming the concept that these REE patterns ($\text{La}{}_{\mathrm{N}}\text{/}\text{Yb}{}_{\mathrm{N}}\text{= 9.5}\text{Eu}\text{/}\text{Eu}{}^1\text{= 0.66}$) represent the upper crustal average. Sm-Nd model ages are variable but <1700 m.y. They reflect derivation from younger elevated erogenic areas subject to Pleistocene glaciation. Although Sm-Nd model ages vary by a factor of two, the REE patterns remain constant. This indicates that processes responsible for formation of the upper crust have produced no secular change in composition since the mid-Proterozoic.     

Large ion lithophile elements in rocks from high-pressure granulite facies terrains

A comparison of K, Rb, Th and U concentrations in granulite facies rocks with those of unmetamorphosed common rock types shows that depletion of these elements in granulites is variable. K/Rb ratios for granulites are generally higher than unmetamorphosed rocks, but K/Rb ratios only reach extreme values when K < 1%. The covariation of K/Rb ratio with K concentration suggests that protolith composition, hence mineralogy, is very important in controlling the degree of Rb depletion in granulites. Felsic granulites exhibiting extreme K/Rb ratios are mainly Archean, reflecting the high abundance of low K felsic rocks in Archean terrains. The Scourian granulites of Scotland all have very high K/Rb ratios and cannot be considered to be representative of granulite facies terrains. It is impossible from this data set to state conclusively whether K is depleted in granulites; K/La ratios of granulites show complete overlap with igneous rocks. Th/U ratios in many granulites are greater than 4, indicating U loss relative to Th. Felsic granulites with low Th/U ratios also have high La/Th ratios, indicating that these granulites have been depleted in Th. The low Th/U ratios of these rocks may reflect retention of Th and U in resistant accessory phases.     

Chemical relationships among irghizites,zhamanshinites, Australasian tektites and Henbury impact glasses

Analytical data for 40 elements are presented for three irghizites (tektite-like forms) and two zhamanshinites (impact glasses) from the Zhamanshin impact structure (Lat. 49°N, Long. 61°E), 200 km north of the Aral Sea, Kazakstahn SSR, USSR. Comparative data for 10 Australasian tektites (3 javanites, 5 australites, an indochinite and a philippinite), Henbury impact glass and parental Henbury subgreywacke sedimentary rocks are also presented.There is close similarity in composition between the three silica-rich samples from Zhamanshin (two irghizites and one zhamanshinite) and the high Mg-Cr tektites from Java, except that the irghizites contain high Ni and Co contamination from the impacting body. The silica-rich zhamanshinite is nearly identical in composition to Henbury impact glass. The refractory trace elements and REE show close similarities among all 17 silica-rich samples, although the tektites show a smaller spread in composition compared with the samples from Zhamanshin. All the silica-rich samples have REE patterns consistent with derivation from terrestrial sedimentary rocks of post-Archean age. All are thoroughly terrestrial. The silica-poor zhamanshinite and irghizite have Al, Na, K, Ni, Cr, Th, U and REE abundances consistent with derivation from a basaltic andesite parent material.The close similarity in chemistry and age between the silica-rich samples from Zhamanshin and Javan tektites from the Australasian strewnfleld warrants a close examination of the Zhamanshin impact structure as a possible source region for the Australasian tektites.